Ops Manager Crash Loop


I tried to create an index on 2 array fields which I realized after I deployed the change that it is computationally not best practice in MongoDB.

The deployment was made on Ops Manager and it caused an error along the lines of “cannot index on parallel arrays” that went into a crash loop.

I could not edit configuration or cancel the operation - I tried dropping the index - even killing the index_creation opId, restoring the database over the configuration with fresh data - removing all of the /data/* but the crash loop continued.

I really think that Ops Manager should have a “cancel deployment” option.
This was at least on a test instance but I will soon create a ticket to MongoDB to fix this as the instance is now un-useable.